One hundred fifty (150) youth (9- 13 year-olds) diagnosed with a primary anxiety disorder will participate in a randomized clinical trial, with 50 cases randomly assigned to Individual Cognitive-Behavioral Therapy (ICBT), 50 cases to Family Cognitive-Behavioral Therapy (FCBT), and 50 cases to an education/support/attention (ESA) control condition. All conditions will be 16 sessions. ICBT and FCBT involve anxiety education, relaxation, relabeling of anxiety-related cognition, problem-solving, exposure, and rehearsal. The ESA condition provides participants with support and attention from a therapist and education about emotions and anxiety. The study will employ multimethod assessment including structured diagnostic interviews, child and parent self-reports, parent and teacher ratings of the child, behavioral observations, and family assessments. Measures will be gathered at pre- and post-treatment, and at one-year follow-up. To ensure that treatment-as-described was provided, treatment integrity will be assessed. Behavioral observations will be taken from videotapes, with two raters (.85 reliability; Kappa) for the child observational system and two for the family coding system. Treatment effects will be analyzed via 3 (conditions; between subjects) X 3 (assessments; within subjects) MANOVA and mixed factorial analysis of variance. Because several dependent measures may be highly interrelated (e.g., self-reports) multivariate analyses of variance will be used for same-method measures (e.g., self-reports). The study will examine maintenance via analyses of one-year follow-up data. The analyses will be conducted for both treatment completers and the intent-to-treat sample. Examinations of clinical significance, predictors of treatment outcome, and differential responsiveness to the two treatments will be conducted.